Where Do Your Competitors Buy Traffic? A Network Detection Guide

You can find out where a competitor buys traffic without any inside information, because every paid channel leaves fingerprints: click IDs and UTM parameters on landing URLs, tracker domains in the redirect chain, official ad libraries on the platforms required to publish them, and observed placements on the networks that are not. Work through those four evidence layers in order — URL, libraries, redirect chain, independent index — and you can usually reconstruct a competitor's channel mix in under an hour, along with a rough read on which channel carries most of their volume.

Method 1: read their landing-page URLs#

The fastest tell is sitting in the address bar. Ad platforms append identifiers to destination URLs so advertisers can attribute clicks, and most advertisers never strip them. Two families matter.

Click IDs are platform-generated and near-impossible to fake accidentally. The stable ones worth memorizing:

UTM parameters are advertiser-written, so they are hints rather than proof — but native buyers habitually set utm_source=taboola, utm_source=outbrain, utm_source=mgid and map network macros (site ID, campaign ID) into utm_campaign or utm_content. A URL like ?utm_source=outbrain&utm_medium=discovery&utm_campaign=us_desktop_v3 tells you the network, the device split, and that they are on at least a third campaign iteration. The click ID glossary entry covers how these identifiers work mechanically.

Where do you get their landing URLs? From their ads (methods 2 and 4), from affiliate-network offer pages, and sometimes from organic sources that preserve tagged URLs — but treat parameter evidence as strongest when you captured the click path yourself.

One caution before you build conclusions on parameters alone: absence proves nothing. Sophisticated advertisers strip or rewrite parameters server-side, and some route every channel through a single tracker that swallows the originals. A clean URL means you move to the next method, not that the competitor buys no traffic.

Method 2: check the official ad libraries#

Three platforms will simply tell you, for free, in about two minutes each:

• Meta Ad Library — search the brand's page; every active Facebook/Instagram ad is listed.
• Google Ads Transparency Center — search the advertiser; shows ads across Google surfaces including YouTube.
• TikTok's Commercial Content Library — covers ads shown in the EEA, useful even for US-focused research since many advertisers run both.

Presence is proof. Absence is weaker evidence — an advertiser can be between flights, or running through a page you did not think to search — but a systematic absence across all three libraries for a brand that is clearly buying traffic somewhere is itself informative: it points you at the channels with no official library, which is exactly where the next two methods live.

Method 3: follow the redirect chain#

Affiliate and native buyers rarely send clicks straight to a landing page. The click passes through one or more tracking hops — an affiliate tracker, a link shortener, sometimes network-side redirectors — before it settles. Each hop is a fingerprint. Tracker domains and their URL patterns identify the tracking software, and the parameters carried across hops frequently name the traffic source outright (?source=, ?site=, publisher and widget IDs), because that is precisely the data the advertiser needs for their own source-level reporting.

Capturing a chain is mundane: open the ad's click URL with your browser's network tab recording, and read the sequence of redirects. What the hops mean and how to interpret them is covered in the redirect chain glossary entry, and OpenAdLibrary records these paths at scale as click-traces — auditable click-to-landing evidence attached to captured ads, with over 1.3 million landing captures resolved to advertisers as of July 2026.

Method 4: search an independent native index#

For native networks the situation inverts: Taboola, Outbrain, MGID, Revcontent, MediaGo and their peers publish no official ad libraries, so the only public evidence of who buys there is independent observation — actually capturing ads from publisher feeds and resolving which advertiser is behind each one.

This is what OpenAdLibrary is built for. Search a competitor's brand or domain in the ad intelligence platform and you get their observed footprint across all 49 indexed networks at once: which networks serve their creatives, in which geos, on which devices, since when, and with how many active variations. Because the index resolves landing pages back to advertisers, it also works in reverse — start from a domain you keep seeing in the wild and identify everything else that advertiser runs. For the placement-level version of this question ("what network served this specific ad I'm looking at?"), use the field guide to identifying the ad network behind any ad.

Observed placements are the strongest evidence in this whole stack: not a parameter someone typed, but the ad physically served on a real page, captured with its creative, network, and destination.

The geo and device dimensions matter more than most researchers expect. A competitor who looks small in your home market can be running their real volume in Germany or Brazil, and a desktop-only footprint versus a mobile-heavy one implies entirely different funnel economics. When you read an advertiser's footprint, read it as a matrix — network by geo by device — rather than a flat list of networks, because the empty cells in that matrix are your least contested entry points.

Method 5: confirm with supply-chain files#

When you have seen a competitor's ad on a specific publisher and want to confirm which pipe carried it, the open-web supply chain is documented in two public files. A publisher's ads.txt lists every company authorized to sell its inventory, with account IDs; a network's sellers.json discloses the entities it sells for. Cross-referencing the two tells you which seller-network relationships could have delivered the impression you saw — useful for distinguishing "bought direct on Taboola" from "arrived through resold demand," a distinction that matters if you plan to buy the same placement yourself. The IAB Tech Lab's ads.txt specification documents the format if you want to read the files raw.

Putting it together: the detection worksheet#

For one competitor, in order:

1. Libraries first (10 min): Meta, Google, TikTok lookups. Record active/inactive and rough creative counts.
2. Native index (10 min): advertiser search across networks; record networks, geos, first-seen dates, variation counts.
3. URL forensics (10 min): collect landing URLs from captured ads; log click IDs and UTM conventions.
4. Redirect chains (15 min): trace two or three ad clicks; identify tracker domains and source parameters.
5. Weight the channels: variation count and creative refresh rate per channel are your volume proxies — a channel with forty fresh variations is carrying real budget; one stale ad is a leftover. To turn footprint into a spend estimate, use the signals framework in estimating competitor native ad spend, and track relative presence over time as a share-of-voice trend.

Re-run the worksheet monthly; channel mixes drift, and the drift is the story. A competitor who quietly shifts variations from Meta toward native over two consecutive months is telling you something about where their blended costs are heading — the kind of read a one-off snapshot can never give you.

What detection can't tell you#

Honest limits: public evidence shows where a competitor buys and roughly how hard, not what they pay or what converts. Spend estimates from footprint are order-of-magnitude tools, not accounting. And a channel's presence proves it works for their offer and economics — your CPCs, funnel, and margins can produce a different answer on identical traffic. Treat the channel map as a prioritized test list, not a verdict: the channels where multiple competitors sustain long-running ads are where your own testing dollars have the best prior.